We propose clinical research to develop and demonstrate high-quality fetal auditory and visual evoked response signals can be reliably recorded during the human pregnancy, and that this information has clinical diagnostic value. To accomplish this task, we will design and construct a SQUID specifically optimized for the recording of fetal magnetoencephalogram (fMEG). We will first validate auditory and visual evoked field recordings and baseline imaging recording through the comparison of fMEG signals compared to direct scalp electrode recording on same patients in early labor. Also, we will establish normal values throughout the pregnancy for auditory and visual evoked fields and baseline MEG through serial recordings on patients starting at 20 weeks gestational age. Finally, to confirm our ability to detect abnormalities prior to birth, we will also conduct auditory and visual evoked field studies and baseline MEG studies on mothers with a prenatal diagnosis of hydrocephalus, neural tube defects, or chromosome abnormalities. In 1988, the National Health Interview Survey estimates the following incidents of neurological disease: cerebral palsy-full-term-2/1,000, preterm-approximately 16/1,000; epilepsy-2/1,000; deafness/hearing loss-15/1,000; blindness- 12/1,000. In 1985, it was estimated that 850,000 children were mentally retarded while 750,000 individuals suffered from cerebral palsy. Neurological and cognitive disorders cost society an estimated 1 billion dollars each year. Inaccessibility has prevented the application of many existing technologies to fetal assessment. Ultrasound and magnetic resonance imaging can detect gross structural neurological diseases but lack the sensitivity to determine functional integrity of the central nervous system. Fetal heart rate monitoring, though sensitive, lacks specificity and may often result in unnecessary premature deliveries. Commercially available low- temperature DC-SQUID sensors appear to have sufficient sensitivity for application in studies of the fetal cortical evoked response. Our group has successfully recorded the auditory evoked field response N1 (200 ms fetal) from fetuses of three patients ranging in gestational age from 33 to 40 weeks. We now propose the design and construction of a large, curved array (850 cm2) engineered to fit the contour of the mother's anterior abdominal wall and to allow non-invasive fetal neurological evaluation.